University of Nebraska - Lincoln University of Nebraska - Lincoln

DigitalCommons@University of Nebraska - Lincoln DigitalCommons@University of Nebraska - Lincoln

Papers in Veterinary and Biomedical Science Veterinary and Biomedical Sciences, Department of

8-1-1997

Overexpression of the D-Alanine Racemase Gene Confers Overexpression of the D-Alanine Racemase Gene Confers Resistance to D-Cycloserine in Resistance to D-Cycloserine in Mycobacterium smegmatis Mycobacterium smegmatis

Nancy E. Caceres University of Nebraska - Lincoln

N. Beth Harris University of Nebraska - Lincoln

James F. Wellehan University of Minnesota, St. Paul, Minnesota

Zhengyu Feng University of Nebraska - Lincoln

Vivek Kapur University of Minnesota, St. Paul, Minnesota, vkapur@psu.edu

See next page for additional authors

Follow this and additional works at: https://digitalcommons.unl.edu/vetscipapers

Part of the Veterinary Medicine Commons

Caceres, Nancy E.; Harris, N. Beth; Wellehan, James F.; Feng, Zhengyu; Kapur, Vivek; and Barletta, Raul G., "Overexpression of the D-Alanine Racemase Gene Confers Resistance to D-Cycloserine in Mycobacterium smegmatis" (1997). Papers in Veterinary and Biomedical Science. 11. https://digitalcommons.unl.edu/vetscipapers/11

This Article is brought to you for free and open access by the Veterinary and Biomedical Sciences, Department of at DigitalCommons@University of Nebraska - Lincoln. It has been accepted for inclusion in Papers in Veterinary and Biomedical Science by an authorized administrator of DigitalCommons@University of Nebraska - Lincoln.

https://digitalcommons.unl.edu/https://digitalcommons.unl.edu/vetscipapershttps://digitalcommons.unl.edu/vetbiomedscihttps://digitalcommons.unl.edu/vetbiomedscihttps://digitalcommons.unl.edu/vetscipapers?utm_source=digitalcommons.unl.edu%2Fvetscipapers%2F11&utm_medium=PDF&utm_campaign=PDFCoverPageshttp://network.bepress.com/hgg/discipline/760?utm_source=digitalcommons.unl.edu%2Fvetscipapers%2F11&utm_medium=PDF&utm_campaign=PDFCoverPageshttps://digitalcommons.unl.edu/vetscipapers/11?utm_source=digitalcommons.unl.edu%2Fvetscipapers%2F11&utm_medium=PDF&utm_campaign=PDFCoverPages

Authors Authors Nancy E. Caceres, N. Beth Harris, James F. Wellehan, Zhengyu Feng, Vivek Kapur, and Raul G. Barletta

This article is available at DigitalCommons@University of Nebraska - Lincoln: https://digitalcommons.unl.edu/vetscipapers/11

https://digitalcommons.unl.edu/vetscipapers/11https://digitalcommons.unl.edu/vetscipapers/11

JOURNAL OF BACTERIOLOGY,0021-9193/97/$04.0010

Aug. 1997, p. 5046–5055 Vol. 179, No. 16

Copyright © 1997, American Society for Microbiology

Overexpression of the D-Alanine Racemase Gene ConfersResistance to D-Cycloserine in Mycobacterium smegmatis

NANCY E. CÁCERES,1 N. BETH HARRIS,1 JAMES F. WELLEHAN,2 ZHENGYU FENG,1

VIVEK KAPUR,2 AND RAÚL G. BARLETTA1,3*

Department of Veterinary and Biomedical Sciences1 and Center for Biotechnology,3

University of Nebraska, Lincoln, Nebraska 68583-0905, and Departmentof Veterinary Pathobiology, University of Minnesota,

St. Paul, Minnesota 551082

Received 21 February 1997/Accepted 11 June 1997

D-Cycloserine is an effective second-line drug against Mycobacterium avium and Mycobacterium tuberculosis.To analyze the genetic determinants of D-cycloserine resistance in mycobacteria, a library of a resistant Myco-bacterium smegmatis mutant was constructed. A resistant clone harboring a recombinant plasmid with a 3.1-kbinsert that contained the glutamate decarboxylase (gadA) and D-alanine racemase (alrA) genes was identified.Subcloning experiments demonstrated that alrA was necessary and sufficient to confer a D-cycloserine resis-tance phenotype. The D-alanine racemase activities of wild-type and recombinant M. smegmatis strains wereinhibited by D-cycloserine in a concentration-dependent manner. The D-cycloserine resistance phenotype in therecombinant clone was due to the overexpression of the wild-type alrA gene in a multicopy vector. Analysis ofa spontaneous resistant mutant also demonstrated overproduction of wild-type AlrA enzyme. Nucleotide se-quence analysis of the overproducing mutant revealed a single transversion (G3T) at the alrA promoter, whichresulted in elevated b-galactosidase reporter gene expression. Furthermore, transformants of Mycobacteriumintracellulare and Mycobacterium bovis BCG carrying the M. smegmatis wild-type alrA gene in a multicopy vectorwere resistant to D-cycloserine, suggesting that AlrA overproduction is a potential mechanism of D-cycloserineresistance in clinical isolates of M. tuberculosis and other pathogenic mycobacteria. In conclusion, these resultsshow that one of the mechanisms of D-cycloserine resistance in M. smegmatis involves the overexpression of thealrA gene due to a promoter-up mutation.

The resurgence of tuberculosis has been characterized bythe emergence of significant numbers of drug-resistant strains.Furthermore, microorganisms of the Mycobacterium aviumcomplex, opportunistic pathogens common in AIDS patients,are inherently resistant to many traditional antimycobacterialagents (20, 23). Hence, the development of novel drugs for thetreatment of atypical infections by M. avium, Mycobacteriumintracellulare, and multiple-drug-resistant Mycobacterium tuber-culosis is urgently needed.

The mycobacterial cell wall is an effective barrier that con-tributes to drug resistance (45). Inhibitors of cell wall biosyn-thesis not only are potential antimycobacterial agents but alsoincrease mycobacterial susceptibility to other antimicrobialagents (36). One inhibitor of cell wall synthesis is D-cycloserine(D-4-amino-isoxazolidone [DCS]), a cyclic structural analog ofD-alanine (31). D-Amino acids, especially D-alanine, D-gluta-mate, and D-aminopimelate, are important components of allbacterial cell walls, including those of mycobacteria. Alanine isusually available as the L stereoisomer, and the conversion toD-alanine by the cytoplasmic enzyme D-alanine racemase (25)is required for the initial step in the alanine branch of pepti-doglycan biosynthesis. D-Alanine is converted to the dipeptideD-alanyl–D-alanine in a reaction catalyzed by D-alanyl:alaninesynthetase (D-alanine ligase [30]). In Escherichia coli, bothD-alanine racemase and D-alanine ligase are targets of DCS(26, 31, 33). Moreover, the biosynthesis of mycolyl-arabinoga-lactan-peptidoglycan complex is inhibited by DCS in M. tuber-

culosis (10), and biochemical studies indicated that D-alanineligase is one of the targets in mycobacteria (11).

DCS is an effective antimycobacterial agent but is rarelyprescribed and used only in combined therapies due to itsadverse effects (21, 22, 54). These side effects are due to bind-ing of DCS to neuronal N-methyl aspartate receptors (44) andinhibition of enzymes that metabolize and synthesize the neu-rotransmitter g-aminobutyric acid (53). Nevertheless, DCS isan excellent candidate for the development of a new genera-tion of antibiotics. Two important considerations predict thatrationally designed derivatives of DCS may be more efficaciousantimicrobial agents. First, DCS targets participate in essentialsteps of cell wall synthesis. Second, DCS resistance has not yetbecome an important clinical problem. Therefore, the identi-fication of DCS targets and the elucidation of the mechanismsleading to DCS resistance may contribute to the developmentof new therapeutics with fewer side effects and mechanisms ofaction which do not favor the emergence of resistance.

Few studies on the mode of action and mechanisms of DCSresistance in mycobacteria have been conducted. David (9)isolated and characterized step-wise DCS-resistant (DCSr)mutants of M. tuberculosis and discovered mutants that showedeither normal or reduced cellular permeability to DCS. It washypothesized that mutants with normal uptake carried muta-tions in the D-alanine ligase gene, but no biochemical or mo-lecular evidence in support of this hypothesis was provided.

Here we describe the first molecular genetic analysis of DCSresistance in mycobacteria, which led to the identification ofone of the DCS targets and resistance mechanisms in Myco-bacterium smegmatis. A spontaneous DCSr mutant strain of M.smegmatis exhibited a promoter-up mutation in the D-alanineracemase gene (alrA) which increased the levels of expression

* Corresponding author. Mailing address: Department of Veteri-nary and Biomedical Sciences, 211 VBS, Fair St. and East CampusLoop, University of Nebraska, Lincoln, NE 68583-0905. Phone: (402)472-8543. Fax: (402) 472-9690. E-mail: braul@crcvms.unl.edu.

5046

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

of the gene and determined a DCSr phenotype. Furthermore,transformants of M. intracellulare and M. bovis BCG with theM. smegmatis alrA gene carried in a multicopy vector had aDCSr phenotype, indicating that a similar mechanism of resis-tance may occur in pathogenic mycobacteria.

MATERIALS AND METHODS

Bacterial strains, plasmids, and culture conditions. Bacterial strains and plas-mids used in this study are listed in Table 1. E. coli strains were grown inLuria-Bertani broth or agar. M. intracellulare strains were grown as previouslydescribed (16). M. smegmatis strains were grown at 37°C with shaking (200 rpm;Innova 4300 incubator shaker; New Brunswick Scientific, Edison, N.J.) inMiddlebrook 7H9 broth (BBL Microbiology Systems, Cockeysville, Md.), Luria-Bertani broth, or minimal medium (55) in the presence of 0.05% Tween 80.Tryptic soy agar base (Difco Laboratories, Detroit, Mich.) was used for growthof M. smegmatis on solid media. Spontaneous DCS-resistant M. smegmatis mu-tants were isolated by plating approximately 5.0 3 109 exponentially growingcells on tryptic soy agar with 500 to 600 mg of DCS ml21. Since DCS is moder-ately unstable, agar plates containing DCS were kept at 4°C for a maximum of 2days. The independent mutants GPM14 and GPM16 were isolated at 500 and600 mg ml21, respectively, from the susceptible parent strain mc2155 (Fig. 1). Todetermine inhibition of colony formation, appropriate dilutions of exponentiallygrowing cells (5.0 3 108 to 1.0 3 109 CFU ml21) were plated in triplicate ontoagar containing a maximum of 1,500 mg of DCS ml21 for M. smegmatis or up to250 mg of DCS ml21 for M. intracellulare. Regression analyses of bacterial titersat different DCS concentrations were conducted by using Proc Reg (SAS Insti-tute, Cary, N.C.). The reduced model for the different bacterial strains was testedagainst the full model for lack of fit for each of the mycobacterial species tested.A significant lack of fit indicated that the responses to the DCS concentrationswere different between the strains. If a significant lack of fit was determined whenall the strains were considered, subsets of strains were also compared.

For D-alanine racemase assays, M. smegmatis cells were grown in minimalmedium (55) to mid-exponential phase (ca. 3.0 3 108 CFU ml21). For b-galactosidase assays, the M. smegmatis strains were grown in Middlebrook 7H9broth to mid-exponential phase (ca. 3.0 3 108 CFU ml21).

Transformation of E. coli and mycobacteria was carried out as describedpreviously (16). E. coli and M. intracellulare transformants were selected at 50 mgof kanamycin ml21. M. smegmatis transformants were selected at either 10 mg ofkanamycin ml21 or 10 mg of kanamycin (Sigma Chemical Co., St. Louis, Mo.)ml21 plus 300 mg of DCS (Sigma or Aldrich Chemical Co., Inc., Milwaukee,Wis.) ml21.

Oligonucleotides, PCR amplifications, and probe labeling. The oligonucleo-tide primers (Ransom Hill Biosciences, Inc., Ramona, Calif.) for PCR amplifi-cation of the D-alanine racemase genes from pBUN66 and GPM14 were JIM-1(59-GNGAYYNYGGRTACACCGAGTTC-39) and JIM-2 (59-CGNCGRCGAGCNNCTCGAAATC-39). The oligonucleotide primer pair for probe labelingby PCR was NAN-1 (59-TCTGCGGCCTCTGGGACAATGGG-39) and NAN-2

FIG. 1. Inhibition of CFU at increasing DCS concentrations for M. smegma-tis mc2155 (open circles), DCS-resistant mutant GPM14 (closed squares), DCS-resistant mutant GPM16 (closed circles), mc2155(pMV262) (open triangles), andmc2155(pBUN19) (closed triangles). The curves were generated from data froma representative experiment. For statistical analysis, performed for at least twoindependent experiments for each strain, see the text.

TABLE 1. Strains and plasmids used in this study

Strain or plasmid Relevant characteristics Source or reference

E. coli DH5a recA lacZDM15 Gibco-BRLM. smegmatis mc2155 High-transformation mutant of M. smegmatis ATCC 607 41M. smegmatis GPM14a M. smegmatis first-step DCSr spontaneous mutant derived from mc2155; overproduces

D-alanine racemaseThis work

M. smegmatis GPM16 M. smegmatis first-step DCSr spontaneous mutant derived from mc2155 This workM. intracellulare mc276 Highly transformable M. avium complex strain W. R. Jacobs, Jr., 16M. bovis BCG French isolate (Pasteur substrain) W. R. Jacobs, Jr.pCV77 Replicating E. coli-Mycobacterium shuttle plasmid; carries cassette of the promoterless

lacZ gene with ribosome-binding site outflanked by transcriptional terminatorsMedImmune Inc.

pMV203 Replicating E. coli-Mycobacterium shuttle plasmid; precursor of pMV262 without Phsp60 MedImmune Inc.pMV262 Replicating E. coli-Mycobacterium shuttle plasmid; carries Phsp60 promoter upstream

from polylinker siteMedImmune Inc., 7

pYUB178 Integration-proficient shuttle cosmid vector; integrates at the attachment site of myco-bacteriophage L5

34

pBUN19 pMV262 with the 3.1-kb insert from GPM16 in the BamHI site This workpBUN25 pMV262 with the 0.9-kb PstI fragment of pBUN19 in the PstI site This workpBUN47D pMV262 with the 2.0-kb PstI fragment of pBUN19 in the PstI site This workpBUN66 Recombinant plasmid isolated from an M. smegmatis mc2155 cosmid library which hy-

bridized with the 3.1-kb insert of pBUN1924, W. R. Jacobs, Jr.,

this workpBUN82 pMV262 with the 1.9-kb ScaI/HindIII fragment of pBUN19 in the PvuII/HindIII site This workpBUN83 pMV262 with the 2.0-kb DraI/ClaI fragment of pBUN19 in the DraIClaI site This workpBUN92 pMV203 with the 2.9-kb EcoRI/EcoRV fragment of pBUN19 in the EcoRI/HpaI site This workpBUN101 pCV77 with the 0.5-kb fragment containing the upstream noncoding region of the alrA

gene from mc2155 inserted at the polylinker siteThis work

pBUN102B pCV77 with the 0.5-kb fragment containing the upstream noncoding region of the alrAgene from GPM14 inserted at the polylinker site

This work

a GPM, Great Plains Mycobacterial Collection.

VOL. 179, 1997 MYCOBACTERIAL D-ALANINE RACEMASE GENE 5047

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

(59-GACACACCTGCCACGGTGCCGAC-39). The amplifications of the up-stream, noncoding regions of the D-alanine racemase genes from mc2155 andGPM14 were done with JIM-2 and JIM-3 (59GTGTGGCGCAACAAAGAG-39). PCR amplifications were carried out with Taq DNA polymerase (FisherScientific Co., Pittsburgh, Pa.) for 30 cycles in a thermal cycler (Perkin-ElmerGeneAmp PCR System 2400; Roche Molecular Systems, Branchburg, N.J.) andrequired 10 mM Tris-HCl (pH 8.3), 2.5 mM MgCl2, 0.2 mM deoxynucleosidetriphosphate, 0.2 mM spermidine, 10% (vol/vol) dimethyl sulfoxide, and 0.1 mgof gelatin ml21 under standard cycling temperatures (hot start at 95°C, denatur-ation at 94°C, annealing at 55°C, and polymerization at 72°C). For radioactivelabeling, 10 cycles were run in the presence of 50 mCi of [a-32P]dATP. TheExpand high-fidelity PCR system (Boehringer Mannheim Biochemicals, India-napolis, Ind.) was used as recommended by the manufacturer.

Nucleic acid manipulations, DNA sequencing, and primer extension analysis.For restriction digestions, ligations, agarose gel electrophoresis, and Southernhybridizations under stringent conditions, standard procedures were followed aspreviously described (40). Chromosomal DNA from M. intracellulare strains, M.paratuberculosis, and M. smegmatis was prepared as described previously (51).Chromosomal DNA from BCG and M. tuberculosis was provided by W. R.Jacobs, Jr., and T. Weisbrod.

Total RNA from mc2155, GPM14, and GPM16 was isolated by following theprocedure described by Bashyam and Tyagi (3). Northern blotting was done aspreviously described (1). Quantification of mRNA was carried out by capturingthe images from the autoradiograms with a Kaiser RS-1 video camera and aNorthern Light model 890 illuminator (Kaiser Optical Systems, Inc., Ann Arbor,Mich.) and analyzing the output with NIH Image 1.49 software. The levels ofRNA were normalized by the amount of rRNA.

Sequencing reactions were carried out with the Taq DyeDeoxy FS terminatorcycle sequencing kit (The Perkin-Elmer Corp., Norwalk, Conn.). The unincor-porated dye terminators and primers were separated from the extension productsby spin column purification (Centri-Sep; Princeton Separations, Adelphia, N.J.).The samples were dried, resuspended in loading buffer, heat denatured, andloaded in ABI model 377 DNA sequencers (Applied Biosystems, Foster City,Calif.). Each template was sequenced in its entirety in both orientations toprevent potential errors in sequencing. DNA sequencing and nucleotide se-quence analyses were performed at the University of Minnesota’s AdvancedGenetic Analysis Center (St. Paul). Protein sequence analysis was performedwith the Genetics Computer Group package (version 8.1), University of Wis-consin (13).

For the assessment of promoter strength by b-galactosidase reporter geneassays, 1.8-kb DNA fragments were amplified with primers JIM-2 and JIM-3, theproducts were digested with ScaI and ClaI, and the resulting 0.5-kb fragmentswere directionally cloned into the ClaI and the blunt-ended PstI sites of pCV77.The synthesis of blunt-ended termini was carried out with Pfu DNA polymeraseas instructed by the supplier (Stratagene, La Jolla, Calif.). All constructs wereverified by sequencing of the relevant regions.

Primer extension analysis of alrA mRNA was carried out as described previ-ously (12). The oligonucleotide NAN-3 (59-ATCGATCACCGTCTGTGCCGACGCC-39) was radiolabeled with [g-32P]ATP by using T4 polynucleotide kinase(Promega). The reactions were extended with Moloney murine leukemia virusreverse transcriptase (Promega). Radioactivity in primer extension bands wasquantified with a PhosphorImager by using ImageQuant software version 3.3(Molecular Dynamics, Sunnyvale, Calif.).

M. smegmatis genomic libraries. An M. smegmatis mc2155 cosmid library waskindly provided by W. R. Jacobs, Jr. For M. smegmatis GPM16 genomic libraries,chromosomal DNA was partially digested with Sau3A, and fragments with anaverage size of 3.0 kb were purified from a 0.8% agarose gel and ligated into theBamHI site of the E. coli-Mycobacterium shuttle plasmid pMV262 (7), whichcarries the kanamycin-resistant marker and the strong promoter Phsp60. Theligation mixture (approximately 1.0 mg of vector DNA) was transformed into M.smegmatis mc2155. For selection of DCS resistance determinants, cells wereplated on 10 mg of kanamycin ml21 plus 300 mg of DCS ml21 and yielded twoDCS-resistant clones. Parallel platings of the transformation mixture on 10 mg ofkanamycin ml21 indicated a transformation efficiency of 4 3 104. Plasmids wereisolated from independent kanamycin-resistant transformants and analyzed in E.coli for the presence of M. smegmatis inserts. This analysis revealed approxi-mately 10% recombinant plasmids. Hence, the two DCS-resistant clones resultedfrom a representative library of ca. 4,000 recombinants.

Preparation of crude cell extracts. Cells were harvested and concentrated20-fold in 50 mM Tris-HCl (pH 8.0). Cells were kept on a salt-ice-water bath andsonicated with a Vibra-Cell model VC600 disrupter (Sonic and Materials, Inc.,Danbury, Conn.). Sonication was carried out for 10 min at 80% power outputand 50% duty cycle, and in the presence of one-third the final volume of type A-5alumina (Sigma). The resulting extracts were centrifuged at 4°C in a JA-17 rotor(Beckman Instruments, Inc., Fullerton, Calif.) for 30 min at 15,000 rpm, dialyzedagainst 50 mM Tris-HCl (pH 8.0), and sterilized by filtration through a 0.22-mm-pore-size filter. The protein concentration was determined by the DC assay(Bio-Rad Laboratories, Richmond, Calif.) as recommended by the manufac-turer.

Enzyme assays. D-Alanine racemase activity in crude extracts was assayed inthe direction of the conversion of L-alanine into D-alanine by a modification ofthe coupled spectrophotometric method described by Wijsman (52). Pilot exper-

iments were performed to determine the amount of each extract and incubationtimes (15 min) resulting in a linear conversion of the substrate into the product.Crude cell extracts were incubated at 37°C in 1.0 ml each of reaction mixturescontaining 50 mM Tris-HCl (pH 8.0), 0.1 mM pyridoxal phosphate (Sigma), and15 mM L-alanine (Sigma). To start the reactions, crude cell extracts were addedto prewarmed mixtures. For inhibition assays, DCS was added at 10, 50, or 100mg ml21. Reactions were terminated by boiling for 10 min. Subsequently, 0.25 mgof D-amino acid oxidase (Boehringer Mannheim), 0.2 mM NADH (Sigma), and10 U of lactate dehydrogenease (Sigma) were added. The coupled reaction wasmeasured by the change in absorbance at 340 nm after overnight incubation at37°C. All controls and samples were measured in triplicate. For the calculationof specific activities, the background change in absorbance (obtained with boil-ing-inactivated extracts processed in an identical manner) was subtracted fromthe change in absorbance obtained with active extracts. From this net absorbancechange (DA340), the specific activity (in micromoles of L-alanine per minute permilligram) was calculated by using the following equation: [DA340/t]/[6.2 3 CP],where t is time (in minutes), 6.2 is the constant used to convert from A340 intomicromoles (in milliliters per micromole, and CP is protein concentration (inmilligrams per milliliter of reaction mixture). No net change in absorbance wasdetected with active extracts when L-alanine was omitted from the reactionmixtures.

The b-galactosidase activity was determined in crude extracts as describedpreviously (29). Pilot experiments were performed for each cell extract to de-termine the optimal incubation time and protein concentration resulting in alinear hydrolysis of b-o-nitrophenylgalactoside. Units of b-galactosidase activityper milligram were calculated as follows: (DA420 of sample 2 DA420 of control)3 380/(t 3 AP), where 380 is the constant used to convert from A420 intob-galactosidase units. t is time (in minutes) at 28°C, and AP is amount of protein(in milligrams) in the reaction mixture.

Nucleotide sequence accession number. Sequence data corresponding to theM. smegmatis D-alanine racemase gene cloned in pBUN19 and flanking se-quences appear in the EMBL/GenBank/DDBJ nucleotide sequence data librar-ies under accession no. U70872.

RESULTS

Cloning and characterization of M. smegmatis DCS resis-tance determinants. DCSr mutants of M. smegmatis were iso-lated, and a genomic library of a mutant strain was constructedin a multicopy plasmid. By using this type of cloning strategy,either wild-type genes producing a DCSr phenotype due to agene dosage effect or genes with dominant mutations encodingproteins which are insensitive to drug inhibition can be iso-lated. However, this strategy does not readily identify muta-tions involved in DCS transport. In this work, two spontaneousDCSr mutants, GPM14 and GPM16, were isolated at 500 and600 mg of DCS ml21, respectively, from the DCS-sensitive(DCSs) strain mc2155 with a frequency of 1.0 3 1029. TheseDCSr strains were identical to the parent strain with respect togeneration time, colony morphology, phage susceptibility, andsusceptibility to antimicrobial agents other than DCS, indicat-ing that these mutants carry a mutation(s) specific for DCSresistance. The mutant strain GPM16 was selected for theconstruction of a genomic library.

The genomic library was constructed in E. coli-Mycobacte-rium shuttle plasmid pMV262, which replicates with a copynumber of 5 to 10 in mycobacteria and carries the kanamycinresistance selection marker (7) (Table 1). The library wastransferred to the DCSs strain mc2155 for the isolation oftransformants resistant to kanamycin and DCS. Plating of arepresentative library (approximately 4,000 recombinants) ledto the isolation of a recombinant plasmid (pBUN19) carryinga 3.1-kb insert. Retransformation of the DCSs strain mc2155with this plasmid resulted in 100% of the transformants dis-playing a DCSr phenotype, indicating that a DCS resistancedeterminant(s) was present in pBUN19.

The DCSr phenotypes of mutants, recombinants, and con-trols were characterized by the inhibition of colony formationat increasing DCS concentrations (Fig. 1). The control group,strains mc2155 and mc2155(pMV262), had approximately thesame susceptibility to DCS (P 5 0.80). Strains GPM14,

5048 CÁCERES ET AL. J. BACTERIOL.

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

GPM16, and mc2155(pBUN19) were significantly more resis-tant than the control group (P , 0.00001).

ORF and subcloning analysis of the DCS resistance deter-minant. Nucleotide sequence analysis of the 3,059-bp DNAinsert from pBUN19 revealed two open reading frames(ORFs) (Fig. 2). ORF1 (1,236 nucleotides) has significant ho-mology with the E. coli glutamate decarboxylase gene, (gadS[28], or gadA [42]). ORF2 (1, 167 nucleotides) encodes a prod-uct with significant homology to D-alanine racemases fromseveral microbial species. The M. smegmatis D-alanine race-mase gene and its immediate flanking regions were sequenced.An initiation codon (ATG) at position 51 and a putative ribo-somal binding site (GAGAT) separated from the initiationcodon by 7 bp were identified. Subcloning experiments wereperformed to further localize the DCSr determinant within the3.1-kb insert of pBUN19. The ScaI/HindIII DNA fragmentsubcloned in pBUN82, which contains the complete D-alanineracemase gene (alrA) but only a truncated glutamate decar-boxylase gene (gadA), was sufficient for a DCSr phenotype(Fig. 2). In contrast, subcloning of the 2.0-kb DraI/ClaI frag-ment in plasmid pBUN83, which contains the gadA gene, gave

a DCSs phenotype. Furthermore, subcloning of DNA frag-ments into pMV262, which split alrA (plasmids pBUN25 andpBUN47D), resulted in DCSs transformants (Fig. 2). Hence,alrA is necessary and sufficient to confer DCSr to mc2155.

The presence of homologous alrA alleles in pathogenic my-cobacteria was tested by Southern analysis with the D-alanineracemase gene amplified by PCR as a probe (Fig. 3). Thisprobe hybridized under stringent conditions with chromosomalDNA from M. smegmatis, M. intracellulare mc276, Mycobacte-rium paratuberculosis ATCC 19698, M. bovis BCG, and M.tuberculosis. Digestion of M. smegmatis mc2155 DNA withBamHI, EcoRI, and HindIII yielded single bands of 7.3, 10.8,and 20 kb, respectively. Digestion with PstI, which cuts oncewithin M. smegmatis alrA, revealed two bands of 6.4 and 1.2 kbin M. smegmatis, single bands of 7.4 and 5.5 in M. intracellulareand M. paratuberculosis, and bands of 9.5, 7.1, and 0.9 kb instrains of the M. tuberculosis complex. The banding pattern ofthe M. smegmatis DNA PstI digest differed from the patternobtained with the recombinant plasmid pBUN19, which pos-sesses a noncontiguous Sau3A DNA fragment (ca. 0.5 kb)upstream from alrA which carries a PstI site.

FIG. 2. Subcloning analysis of pBUN19 and the DCS-resistant phenotype. The position of the Phsp60 promoter in vector pMV262 is indicated. M. smegmatissequences (open boxes) and the locations of ORFs are shown. The DCS resistance phenotypes of corresponding subclones are indicated by plus (resistant) and minus(sensitive) signs.

VOL. 179, 1997 MYCOBACTERIAL D-ALANINE RACEMASE GENE 5049

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

Sequence analysis of the M. smegmatis D-alanine racemase(alrA) gene. The alignment of the inferred amino acid se-quences of D-alanine racemases from several bacterial speciesis displayed in Fig. 4. The predicted 41.0-kDa polypeptidedisplayed 66% amino acid identity to the homologous pre-dicted polypeptides in M. leprae (17) (accession no. U00020)and M. tuberculosis (35) (accession no. Z77165). The M. smeg-matis polypeptide has approximately 35% identity to the D-alanine racemase isozymes from Bacillus stearothermophilus (2,43), Bacillus subtilis (14), E. coli (5, 27), and Salmonella typhi-murium (49, 50) and the putative enzyme from Haemophilusinfluenzae (15). Multiple amino acid sequence alignmentsshowed highly conserved domains: the amino acid sequenceA66VVKANAYGHG76 in the consensus, which contributes tothe pyridoxal phosphate binding domain in the active site (18),and the conserved lysine which covalently binds pyridoxalphosphate in the catalytic cycle (2, 39). All the mycobacterialD-alanine racemases display a two-domain structure, as ob-served in the isozymes from S. typhimurium (19). In this largeralignment, the consensus at the hinge region that links the twodomains is defined as V302-YG--W308.

Analysis of D-alanine racemase activity and the inhibitoryeffect of DCS in M. smegmatis. We determined the D-alanineracemase specific activity of cell extracts from the parent strain,i.e., mc2155, mutant GPM16, and recombinant strain mc2155(pBUN19) (Fig. 5). The D-alanine racemase activity in eachcell extract was inhibited by DCS in a concentration-dependentmanner. Degrees of inhibition by DCS were similar for allthese strains (data not shown). These results confirmed thatD-alanine racemase is one of the drug targets. Surprisingly, themutant GPM16 displayed levels of D-alanine racemase and apattern of DCS inhibition similar to that of the wild-type strain.Therefore, its resistance phenotype appears to be due to amutation in a separate DCS resistance determinant, distinctfrom the D-alanine racemase gene. In contrast, the recombi-nant strain had a 15-fold-higher specific activity. We hypothe-sized that the resistance phenotype in the recombinant strainwas the consequence of elevated expression of the wild-type

D-alanine racemase gene harbored in the multicopy vector. Toverify this hypothesis, a cosmid library of the wild-type strainmc2155 was screened with the objective of identifying andanalyzing the D-alanine racemase gene. The recombinant plas-mid pBUN66 was isolated by colony hybridization with theDraI-HindIII fragment from pBUN19 as a probe (see Materi-als and Methods). The wild-type D-alanine racemase nucle-otide gene and flanking sequences in pBUN66 were PCRamplified and sequenced. The D-alanine racemase gene inpBUN66 was identical to the one cloned in pBUN19, confirm-ing that the mutant GPM16 had a wild-type D-alanine race-mase gene. Therefore, this result together with the biochemicaldata strongly suggests that the DCS-resistant phenotype of therecombinant strain mc2155(pBUN19) was due to an overex-pression of the D-alanine racemase gene and not a mutation inthe structural gene. Furthermore, the recombinant plasmidpBUN92 (Fig. 2), which does not carry the Phsp60 promoter,exhibits a DCSr phenotype as well. Hence, multiple copies ofthe wild-type D-alanine racemase gene appear to cause theDCSr phenotype of the recombinant strain. In agreement withthe E. coli (5) (accession no. U00006) and M. leprae (17) (ac-cession no. U00020) designations, we propose the designationalrA for the M. smegmatis D-alanine racemase gene, which isconstitutively expressed (6).

Overproduction of D-alanine racemase is also a mechanismof resistance in spontaneous M. smegmatis DCSr mutants.Since the observation of the overproduction of D-alanine race-mase in the recombinant strain mc2155(pBUN19) was the re-sult of a laboratory manipulation, we analyzed four additionalDCSr independent mutants derived from mc2155. The crudeextract of one of these mutants, GPM14, displayed levels ofD-alanine racemase specific activity similar to those displayedby the recombinant strain mc2155(pBUN19) (Fig. 5). Thespontaneous DCSr mutant GPM14 exhibited approximately20-fold-greater D-alanine racemase specific activity than theDCSs strain mc2155. The D-alanine racemase activity ofGPM14 was inhibited by DCS in a fashion similar to that of thewild-type enzyme. In agreement with the enzyme activity as-says, Northern blot analysis demonstrated a 30-fold overex-pression of D-alanine racemase alrA mRNA in GPM14 com-pared to that in strain mc2155 (data not shown).

A promoter-up mutation leads to the overexpression of theD-alanine racemase (alrA) gene in M. smegmatis DCSr mutantGPM14. The D-alanine racemase allele from GPM14 was PCRamplified and sequenced. Nucleotide sequence analysis did notshow any mutation within the structural gene but revealed asingle base change, T for G, in the upstream noncoding re-gions. The identification of the DCSr mutant GPM14, whichpossessed an elevated level of D-alanine racemase specific ac-tivity but an unaltered D-alanine racemase structural gene,established solid evidence that overproduction of this enzymecould occur by a natural mechanism.

Mapping of the mRNA start site in mc2155 and GPM14 wascarried out by primer extension analysis (Fig. 6). In each case,two start sites, one nucleotide apart, were found 13 and 14 bpupstream from the amino acid start codon, with the product ofthe shorter sequence being more abundant. The point muta-tion in GPM14 was located within the putative 210 promoter

FIG. 3. (A) Southern blot of total DNA from M. smegmatis digested withBamHI (lane 1), EcoRI (lane 2), HindIII (lane 3), and PstI (lane 4). (B) Southernblot of total DNA from different mycobacterial species digested with PstI: M.intracellulare mc276 (lane 1), M. paratuberculosis ATCC 19698 (lane 2), M. bovisBCG-Pasteur (lane 3), and M. tuberculosis H37Ra (lane 4), H37Rv (lane 5), andErdman (lane 6). Blots were hybridized with the radiolabeled 1,092-bp PCRfragment from pBUN19 under stringent conditions as indicated in Materials andMethods.

FIG. 4. Multiple sequence alignment of bacterial D-alanine racemases that exhibit similarity to the M. smegmatis ORF2 DCS resistance determinant from pBUN19(Msmegm). Black boxes indicate complete identity, and shaded boxes indicate conservative amino acid substitutions. Sequences were obtained from GenBank. Thesequences are from various organisms as follows: Bsth-Cat, B. stearothermophilus (catabolic isozyme; accession no. M19142 [43]); Bsubt, B. subtilis (accession no.M16207 [14]); Mleprae, M. leprae (accession no. U00020 [17]); Mtb, M. tuberculosis (accession no. Z77165 [35]); Ecoli-Cat, E. coli (catabolic isozyme; accession no.L02948 [27]); Styph-Cat, S. typhimurium (catabolic isozyme; accession no. K02119 [49]); Styph-Bio, S. typhimurium (biosynthetic isozyme; accession no. M12847 [18]);Ecoli-Bio, E. coli (biosynthetic isozyme; accession no. U00006 [5]); and Hinf-Bio, H. influenzae (biosynthetic isozyme; accession no. L46206 [15]).

5050 CÁCERES ET AL. J. BACTERIOL.

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

VOL. 179, 1997 MYCOBACTERIAL D-ALANINE RACEMASE GENE 5051

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

box at position 213 in the short transcript. Quantification ofradioactivity in the primer extension bands demonstrated thatboth the shorter and longer transcripts of GPM14 (DCSr) wereoverproduced approximately 15-fold with respect to those inmc2155 (DCSs).

To assess whether the point mutation in the promoter regionof the D-alanine racemase gene of mutant strain GPM14 wasresponsible for changes in the promoter strength, transcrip-tional fusions to a reporter gene were performed. The ScaI-ClaI fragments (Fig. 2), which included approximately 500 bpof the noncoding region of the D-alanine racemase gene fromwild-type (mc2155) and mutant (GPM14) strains, were sub-cloned into the promoter-probe vector pCV77 (Table 1) up-stream from the promoterless lacZ gene. These promoter con-structs were transformed into wild-type M. smegmatis mc2155,and b-galactosidase activities were measured. The expressionfrom the construct containing the GPM14 alrA promoter wasapproximately 50-fold higher than the expression from therespective mc2155 promoter. These results (Table 2) con-firmed that a point mutation, which increases the %AT of theputative 210 box of the alrA gene, led to a significant increasein the level of gene expression.

Overexpression of the M. smegmatis D-alanine racemase(alrA) gene from a multicopy plasmid in M. intracellulare andM. bovis BCG leads to a DCS-resistant phenotype. To deter-mine if the overproduction of D-alanine racemase could alsoconfer a DCSr phenotype in pathogenic mycobacteria, recom-binant plasmid pBUN19 was electrotransformed into M. intra-cellulare mc276 and M. bovis BCG. The susceptibilities of thewild-type and transformant strains to DCS are depicted in Fig.7. These mycobacterial species are inherently more susceptibleto DCS than M. smegmatis. As illustrated, the M. intracellularestrain mc276(pBUN19) is more resistant to DCS than the par-ent strain transformed with plasmid vector pMV262 (P ,0.01). Colony morphology changes were noticed at DCS con-

centrations higher than 40 mg ml21 in M. intracellulare. Asimilar profile was obtained with M. bovis BCG, where thetransformant with plasmid pBUN19 was more resistant thanthe control strain carrying cloning vector pMV262 (P , 0.05).Taken together, these data indicate that multiple copies of awild-type D-alanine racemase gene can confer a DCS-resistantphenotype in M. intracellulare and M. bovis BCG.

DISCUSSION

In this study, we identified a gene from M. smegmatis thatconfers a DCS-resistant phenotype to a wild-type DCS-sensi-tive host when cloned into a multicopy vector. This representsthe first molecular genetic analysis of DCS targets in mycobac-teria. The mechanisms of DCS resistance in mycobacteria havenot been thoroughly investigated. Since DCS can inhibit sev-eral pyridoxal phosphate enzymes (31), the existence of morethan one mechanism leading to DCS resistance is likely. Inaddition, multiple-step DCSr mutants of mycobacteria wereisolated in previous studies (46) as well as in our laboratory.The mechanisms relevant to DCS resistance in mycobacteriaare presented in the model depicted in Fig. 8. DCS may enterthe mycobacterial cell by either diffusion or uptake via a spe-cific transporter (48). Mutations in the transporter gene reduc-ing DCS binding or eliminating the transporter from the cellsurface may lead to reduced DCS uptake and increased DCSresistance. Similarly, a mutation in a gene coding for an effluxpump may lead to higher affinity for DCS, with the concomi-tant expulsion of the drug. Alternatively, it is possible that amutational change could increase the affinity or the levels of adrug-detoxifying enzyme which would derivatize or hydrolyzeDCS. The overproduction of a protein target(s) could lead toincreased resistance due to the sequestration or removal of the

FIG. 5. Analysis of M. smegmatis D-alanine racemase activities. Enzyme ac-tivity was determined in cell extracts from cells grown in minimal medium (55)without DCS. Specific activities are expressed as micromoles of L-alanine perminute per milligram (means 6 standard deviations of triplicate measurements).

FIG. 6. Primer extension analysis of the alrA transcript. Total RNA (50 mg)from mc2155 or GPM14 cells was annealed to an oligonucleotide of the alrA geneand extended as described in Materials and Methods. Lanes A, C, G, and Tdisplay a dideoxy sequencing ladder of the wild-type alrA gene generated withthe same oligonucleotide primer. Nucleotides at the two start sites are indicatedby asterisks. The target site for the mutation in the promoter region is boxed.

5052 CÁCERES ET AL. J. BACTERIOL.

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

drug by the excess target protein. The latter mechanism wouldeffectively reduce the free intracellular DCS concentration,which would also protect other potential protein targets fromdrug inhibition. If DCS interacts with more than one targetprotein (such as D-alanine racemase and D-alanine ligase), it isunlikely that mutations in one of the structural genes alonewould lead to primary DCS resistance. In this study, we pre-sented evidence that one of the mechanisms of DCS resistancein M. smegmatis involves the overexpression of the alrA genedue to a promoter-up mutation. In agreement with our find-ings, a similar mechanism of DCSr involving the overproduc-tion of either D-alanine racemase or D-alanine ligase enzymeactivities, possibly mediated by wild-type products, was de-scribed for streptococci (38).

We have shown that D-alanine racemase is a target of DCSand that a natural mechanism of resistance is the overproduc-tion of the enzyme due to a promoter-up mutation. An inter-esting finding was that the original DCSr mutant, GPM16, usedas a source of DNA for the cloning experiments did not possesselevated levels of D-alanine racemase. Hence, the resistancemechanism in GPM16 and three other mutants is not relatedto the D-alanine racemase determinant. We are currently in-vestigating another DCSr clone carrying a different DCS resis-tance determinant (6).

Soon after the introduction of DCS as an antituberculosisdrug about 40 years ago, DCSr clinical isolates of M. tubercu-losis were readily isolated (8). Since those first trials, DCS hasnot been administered frequently or alone due to its toxicity inpatients. As a consequence, the current panel of DCSr clinicalisolates of pathogenic mycobacteria is limited and may not berepresentative of the situation that would arise in therapy withDCS. Once we identified a DCSr determinant in M. smegmatisand showed its conservation within the genus, we testedwhether a similar mechanism of DCS resistance could occur inmembers of the pathogenic mycobacterial groups by trans-forming M. intracellulare and M. bovis BCG with a multicopyplasmid carrying the M. smegmatis alrA gene. DCSr transfor-mants were obtained, suggesting that AlrA overproduction is apotential mechanism of DCS resistance in clinical isolates ofM. tuberculosis and other pathogenic mycobacteria. Further-more, the homology between M. smegmatis AlrA and the M.leprae and M. tuberculosis counterparts suggests a potentialutility of M. smegmatis as a surrogate host for the study of theAlrA enzymes from pathogenic species.

Since racemases are absent from mammalian cells, thesebacterial enzymes are excellent targets for antibiotic develop-ment (47). Several compounds were developed but were neveradvanced into clinical use due to their toxicity in humans (32).

FIG. 7. Inhibition of CFU at increasing DCS concentrations. (Left) Comparison of M. intracellulare strains mc276(pMV262) (open triangles) and mc276(pBUN19)(closed triangles). (Right) Comparison of M. bovis BCG(pMV262) (open circles) and BCG(pBUN19) (closed circles). The asterisks indicate the concentrations at whichchanges in colony morphology (from opaque white domed to transparent flat colonies) were observed. For analysis of statistical significance, see the text.

TABLE 2. Expression of b-galactosidase activity in recombinant M. smegmatis carrying transcriptional fusions to alrA upstream sequences

Source of alrAupstream sequence Plasmid Relevant alrA upstream sequence

a Sp act ofb-galactosidaseb

None pCV77 None 0.42 6 0.01mc2155 pBUN101 GGGACAATGGGCGCCGGAGATTATGACGATG 7.92 6 0.01GPM14 pBUN102B GGTACAATGGGCGCCGGAGATTATGACGATG (3.6 6 0.2) 3 102

a The putative 210 box, Shine-Dalgarno sequence, and ATG start codon for the alrA gene are underlined; the T-for-G transversion in the upstream noncodingsequences and the transcriptional start site are shown in boldface type. The transversion increases the similarity of the putative 210 box with the consensus E. coli Es70

promoter (37) and the general consensus established for M. smegmatis and M. tuberculosis promoters (4).b The specific activities of b-galactosidase from M. smegmatis mc2155 transformants carrying promoter constructs are expressed in units/milligram of protein

(means 6 standard deviations of triplicate measurements).

VOL. 179, 1997 MYCOBACTERIAL D-ALANINE RACEMASE GENE 5053

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

The D-alanine racemase gene of mycobacteria may be a pow-erful tool for the rational design of effective and less toxic DCSderivatives and D-alanine analogs against M. tuberculosis andother pathogenic mycobacteria. A structure-based layout inwhich computational and crystallography methods are com-bined would advance the rational design of a new generation ofdrugs. A key ingredient to the success of this strategy is theidentification of the DCS lethal target(s) in pathogenic myco-bacteria whose inhibition leads to cell death.

ACKNOWLEDGMENTS

This research was supported by grants from the University of Ne-braska Center for Biotechnology, the Department of Veterinary andBiomedical Sciences of the University of Nebraska, Layman Fundaward, a Nebraska Agriculture Experiment Station InterdisciplinaryResearch award, a subcontract to the Texas Agriculture ExperimentStation Texas Cattle and Deer Tuberculosis Management Plan, andUSDA Department of Agriculture Cooperative State Research Ser-vice Projects NEB 14-077 and NEB-090 and was supported in part byresearch award AI40365 from the National Institutes of Health (toV.K.).

We thank B. C. Jong, N. Le, S. L. Williams, and C. Yost for help inthe experiments. We thank H. Hoff for assistance with the GeneticsComputer Group program. We thank W. R. Jacobs, Jr., and T. Weis-brod for BCG and M. tuberculosis DNA, as well as for the M. smegmatiscosmid library. The assistance of R. Wills with statistical analysis isgreatly appreciated. We thank MedImmune Inc. for the vectorspMV203, pMV262, and pCV77. We acknowledge L. G. Adams, R.Banerjee, L. Bermudez, and J. Cirillo for critical review of the manu-script.

REFERENCES

1. Ausubel, F. M., R. Brent, R. E. Kingston, D. D. Moore, J. G. Seidman, J. A.Smith, and K. Struhl (ed.). 1989. Current protocols in molecular biology, p.1–8. Greene Publishing Associates and Wiley-Interscience, New York, N.Y.

2. Badet, B., K. Inagaki, K. Soda, and C. T. Walsh. 1986. Time-dependentinhibition of Bacillus stearothermophilus alanine racemase by (1-aminoeth-yl)phosphonate isomers by isomerization to noncovalent slowly dissociatingenzyme-(1-aminoethyl)phosphonate complexes. Biochemistry 25:3275–3282.

3. Bashyam, M. D., and A. K. Tyagi. 1994. An efficient and high-yieldingmethod for isolation of RNA from mycobacteria. BioTechniques 17:834–835.

4. Bashyam, M. D., D. Kaushal, S. K. DasGupta, and A. K. Tyagi. 1996. A studyof the mycobacterial transcriptional apparatus: identification of novel fea-tures in promoter elements. J. Bacteriol. 178:4847–4853.

5. Blattner, F. R., V. Burland, G. Plunkett III, H. J. Sofia, and D. L. Daniels.1993. Analysis of the Escherichia coli genome. IV. DNA sequence of theregion from 89.2 to 92.8 minutes. Nucleic Acids Res. 21:5408–5417.

6. Cáceres, N. E., N. B. Harris, and R. G. Barletta. Unpublished data.

7. Connell, N. D., E. Medina-Acosta, W. R. McMaster, B. R. Bloom, and D. G.Russell. 1993. Effective immunization against cutaneous leishmaniasis withrecombinant bacille Calmette-Guerin expressing the Leishmania surfaceprotein gp63. Proc. Natl. Acad. Sci. USA 90:11473–11477.

8. Cummings, M. M. 1956. Cycloserine: resistance data, p. 377. In Transactionsof the 15th Conference on the Chemotherapy of Tuberculosis. VeteransAdministration-Armed Forces, Washington, D.C.

9. David, H. L. 1971. Resistance to D-cycloserine in the tubercle bacilli: muta-tion rate and transport of alanine in parental cells and drug-resistant mu-tants. Appl. Microbiol. 21:888–892.

10. David, H. L., D. S. Goldman, and K. Takayama. 1970. Inhibition of thesynthesis of wax D peptidoglycolipid of Mycobacterium tuberculosis by D-cycloserine. Infect. Immun. 1:74–77.

11. David, H. L., K. Takayama, and D. S. Goldman. 1969. Susceptibility ofmycobacterial D-alanyl–D-alanine synthetase to D-cycloserine. Am. Rev. Re-spir. Dis. 100:579–581.

12. Davis, L. G., W. M. Kuehl, and J. F. Battey (ed.). 1994. Basic methods inmolecular biology, 2nd ed., p. 378–383. Appleton & Lange, Norwalk, Conn.

13. Devereux, J., P. Haeberli, and O. Smithies. 1984. A comprehensive set ofsequence analysis programs for the VAX. Nucleic Acids Res. 12:387–395.

14. Ferrari, E., D. J. Henner, and N. Y. Yang. 1985. Isolation of an alanineracemase gene from Bacillus subtilis and its use for plasmid maintenance inB. subtilis. Bio/Technology 3:1003–1007.

15. Fleischmann, R. D., M. D. Adams, O. White, R. A. Clayton, E. F. Kirkness,A. R. Kerlavage, C. J. Bult, J.-F. Tomb, B. A. Dougherty, J. M. Merrick, K.McKenney, G. Sutton, W. FitzHugh, C. A. Fields, J. D. Gocayne, J. D. Scott,R. Shirley, L.-I. Liu, A. Glodek, J. M. Kelley, J. F. Weidman, C. A. Phillips,T. Spriggs, E. Hedblom, M. D. Cotton, T. R. Utterback, M. C. Hanna, D. T.Nguyen, D. M. Saudek, R. C. Brandon, L. D. Fine, J. L. Fritchman, J. L.Fuhrmann, N. S. M. Geoghagen, C. L. Gnehm, L. A. McDonald, K. V. Small,C. M. Fraser, H. O. Smith, and J. C. Venter. 1995. Whole-genome randomsequencing and assembly of Haemophilus influenzae Rd. Science 269:496–512.

16. Foley-Thomas, E. M., D. L. Whipple, L. B. Bermudez, and R. G. Barletta.1995. Phage infection, transfection and transformation of Mycobacteriumavium complex and Mycobacterium paratuberculosis. Microbiology 141:1173–1181.

17. Fsihi, H., and S. T. Cole. 1995. The Mycobacterium leprae genome: systematicsequence analysis identifies key catabolic enzymes, ATP-dependent trans-port systems and a novel polA locus associated with genomic variability. Mol.Microbiol. 16:909–919.

18. Galakatos, N. G., E. Daub, D. Botstein, and C. T. Walsh. 1986. Biosyntheticalr alanine racemase from Salmonella typhimurium: DNA and protein se-quence determination. Biochemistry 25:3255–3260.

19. Galakatos, N. G., and C. T. Walsh. 1987. Specific proteolysis of nativealanine racemases from Salmonella typhimurium: identification of the cleav-age site and characterization of the clipped two-domain proteins. Biochem-istry 26:8475–8480.

20. Good, R. C. 1985. Opportunistic pathogens in the genus Mycobacterium.Annu. Rev. Microbiol. 39:347–369.

21. Heifets, L. B., and M. D. Iseman. 1991. Individualized therapy versus stan-dard regimens in the treatment of Mycobacterium avium infections. Am. Rev.Respir. Dis. 144:1–2.

22. Helmy, B. 1970. Side effects of cycloserine. Scand. J. Respir. Dis. 71S:220–225.

23. Inderlied, C. B., C. A. Kemper, and L. E. M. Bermudez. 1993. The Myco-bacterium avium complex. Clin. Microbiol. Rev. 6:266–310.

24. Jacobs, W. R., Jr., K. V. Ganjam, J. D. Cirillo, L. Pascopella, S. B. Snapper,R. A. Udani, W. Jones, R. G. Barletta, and B. R. Bloom. 1991. Geneticsystems for the mycobacteria. Methods Enzymol. 204:537–555.

25. Julius, M., C. A. Free, and G. T. Barry. 1970. Alanine racemase (Pseudo-monas). Methods Enzymol. XVIIA:171–176.

26. Lambert, M. P., and F. C. Neuhaus. 1972. Mechanism of D-cycloserineaction: alanine racemase from Escherichia coli W. J. Bacteriol. 110:978–987.

27. Łobocka, M., J. Hennig, J. Wild, and T. Kłopotowski. 1994. Organization andexpression of the Escherichia coli K-12 dad operon encoding the smallersubunit of D-amino acid dehydrogenase and the catabolic alanine racemase.J. Bacteriol. 176:1500–1510.

28. Maras, B., G. Sweeney, D. Barra, F. Bossa, and R. A. John. 1992. The aminoacid sequence of glutamate decarboxylase from Escherichia coli. Evolution-ary relationship between mammalian and bacterial enzymes. Eur. J. Bio-chem. 204:93–98.

29. Miller, J. H. 1972. Experiments in molecular genetics, p. 352–355. ColdSpring Harbor Laboratory, Cold Spring Harbor, N.Y.

30. Neuhaus, F. C. 1962. The enzymatic synthesis of D-alanyl–D-alanine. I. Pu-rification and properties of D-alanyl–D-alanine synthetase. J. Biol. Chem.237:778–786.

31. Neuhaus, F. C. 1967. D-Cycloserine and O-carbamyl-D-serine, p. 40–83. In D.Gottlieb and P. L. Shaw (ed.), Antibiotics, vol. 1. Mechanisms of action.Springer-Verlag, Heidelberg, Germany.

32. Neuhaus, F. C., and W. P. Hammes. 1981. Inhibition of cell wall biosynthesisby analogs of alanine. Pharmacol. Ther. 14:265–319.

FIG. 8. Model of DCS resistance mechanisms in mycobacteria. There arefour proposed mechanisms of resistance: impairments in the DCS transporter,acquisition or overproduction of an efflux pump, enzyme target overproduction,and acquisition or overproduction of a DCS-detoxifying enzyme.

5054 CÁCERES ET AL. J. BACTERIOL.

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

33. Neuhaus, F. C., and J. L. Lynch. 1964. The enzymatic synthesis of D-alanyl–D-alanine. III. On the inhibition of D-alanyl–D-alanine synthetase by theantibiotic D-cycloserine. Biochemistry 3:471–480.

34. Pascopella, L., F. M. Collins, J. M. Martin, M. H. Lee, G. F. Hatfull, C. K.Stover, B. R. Bloom, and W. R. Jacobs, Jr. 1994. Use of in vivo complemen-tation in Mycobacterium tuberculosis to identify a genomic fragment associ-ated with virulence. Infect. Immun. 62:1313–1319.

35. Philipp, W. J., S. Poulet, K. Eiglmeier, L. Pascopella, V. Balasubramanian,B. Heym, S. Bergh, B. R. Bloom, W. R. Jacobs, Jr., and S. T. Cole. 1996. Anintegrated map of the genome of the tubercle bacillus, Mycobacterium tuber-culosis H37Rv, and comparison with Mycobacterium leprae. Proc. Natl. Acad.Sci. USA 93:3132–3137.

36. Rastogi, N., K. S. Goh, and H. L. David. 1990. Enhancement of drug sus-ceptibility of Mycobacterium avium by inhibitors of cell envelope synthesis.Antimicrob. Agents Chemother. 34:759–764.

37. Record, M. T., Jr., W. S. Reznikoff, M. L. Craig, K. L. McQuade, and P. J.Schlax. 1996. Escherichia coli RNA polymerase (Es70), promoters, and thekinetics of the steps of transcription initiation, p. 792–820. In F. C. Nei-dhardt, R. Curtiss III, J. L. Ingraham, E. C. C. Lin, K. B. Low, B. Magasanik,W. S. Reznikoff, M. Riley, M. Schaechter, and H. E. Umbarger (ed.), Esch-erichia coli and Salmonella: cellular and molecular biology, 2nd ed., vol. 1.American Society for Microbiology, Washington, D.C.

38. Reitz, R. H., H. D. Slade, and F. C. Neuhaus. 1967. The biochemical mech-anisms of resistance by streptococci to the antibiotics D-cycloserine andO-carbamyl-D-serine. Biochemistry 6:2561–2570.

39. Roise, D., K. Soda, T. Yagi, and C. T. Walsh. 1984. Inactivation of thePseudomonas striata broad specificity amino acid racemase by D and L iso-mers of beta-substituted alanines: kinetics, stoichiometry, active site peptide,and mechanistic studies. Biochemistry 23:5195–5201.

40. Sambrook, J., E. F. Fritsch, and T. Maniatis. 1989. Molecular cloning: alaboratory manual, 2nd ed. Cold Spring Harbor Laboratory, Cold SpringHarbor, N.Y.

41. Snapper, S. B., R. E. Melton, S. Mustafa, T. Kieser, and W. R. Jacobs, Jr.1990. Isolation and characterization of efficient plasmid transformation mu-tants of Mycobacterium smegmatis. Mol. Microbiol. 4:1911–1919.

42. Sofia, H. J., V. Burland, D. L. Daniels, G. Plunkett, and F. R. Blattner. 1994.Analysis of the Escherichia coli genome. V. DNA sequence of the regionfrom 76.0 to 81.5 minutes. Nucleic Acids Res. 22:2576–2586.

43. Tanizawa, K., A. Ohshima, A. Scheidegger, K. Inagaki, H. Tanaka, and K.Soda. 1988. Thermostable alanine racemase from Bacillus stearothermophi-lus: DNA and protein sequence determination and secondary structure pre-diction. Biochemistry 27:1311–1316.

44. Thompson, L. T., J. R. Moskal, and J. F. Disterhoft. 1992. Hippocampus-dependent learning facilitated by a monoclonal antibody or D-cycloserine.Nature 359:638–641.

45. Trias, J., and R. Benz. 1994. Permeability of the cell wall of Mycobacteriumsmegmatis. Mol. Microbiol. 14:283–290.

46. Tsukamura, M., Y. Noda, M. Hayashi, and M. Yamamoto. 1959. A geneticstudy on the cycloserine-resistance of Mycobacterium tuberculosis var. homi-nis. Jpn. J. Microbiol. 3:1–8.

47. Wang, E., and C. Walsh. 1978. Suicide substrates for the alanine racemase ofEscherichia coli B. Biochemistry 17:1313–1321.

48. Wargel, R. J., C. A. Shadur, and F. C. Neuhaus. 1970. Mechanism of D-cycloserine action: transport systems for D-alanine, D-cycloserine, L-alanine,and glycine. J. Bacteriol. 103:778–788.

49. Wasserman, S. A., E. Daub, P. Grishif, D. Botstein, and C. T. Walsh. 1984.Catabolic alanine racemase from Salmonella typhimurium: DNA sequence,enzyme purification, and characterization. Biochemistry 23:5182–5187.

50. Wasserman, S. A., C. T. Walsh, and D. Botstein. 1983. Two alanine racemasegenes in Salmonella typhimurium that differ in structure and function. J.Bacteriol. 153:1439–1450.

51. Whipple, D. L., R. B. Le Febvre, R. E. Andrews, Jr., and A. B. Thiermann.1987. Isolation and analysis of restriction endonuclease digestive patterns ofchromosomal DNA from Mycobacterium paratuberculosis and other Myco-bacterium species. J. Clin. Microbiol. 25:1511–1515.

52. Wijsman, H. J. W. 1972. The characterization of an alanine racemase mutantof Escherichia coli. Genet. Res. 20:269–277.

53. Wood, J. D., S. J. Peesker, D. K. J. Gorecki, and D. Tsui. 1978. Effect ofL-cycloserine on brain GABA metabolism. Can. J. Physiol. Pharmacol. 52:62–68.

54. Yew, W. W., C. F. Wong, P. C. Wong, J. Lee, and C. H. Chau. 1993. Adverseneurological reactions in patients with multidrug-resistant pulmonary tuber-culosis after coadministration of cycloserine and ofloxacin. Clin. Infect. Dis.17:288–289.

55. Zygmunt, W. A. 1963. Antagonism of D-cycloserine inhibition of mycobac-terial growth by D-alanine. J. Bacteriol. 85:1217–1220.

VOL. 179, 1997 MYCOBACTERIAL D-ALANINE RACEMASE GENE 5055

at UN

IV O

F N

EB

RA

SK

A-LIN

CO

LN on S

eptember 5, 2007

jb.asm.org

Dow

nloaded from

http://jb.asm.org

Overexpression of the D-Alanine Racemase Gene Confers Resistance to D-Cycloserine in Mycobacterium smegmatisAuthors

tmp.1189005198.pdf.g7bg6

Text2: